Silicone rubber materials are commonly used as insulation materials under high-pressure conditions due to their excellent insulation properties. The breakdown field strength is an important electrical performance index, and there is a complex nonlinear relationship between the breakdown field strength and the material formula. Based on this, an efficient evaluation model based on genetic algorithm (GA) optimized extreme gradient boosting (XGBoost) algorithm is proposed. The model combines GA and XGBoost, and uses temperature, relative content of masterbatch, diameter of Al(OH)3 micropowder, relative content of Al(OH)3 and thickness as inputs to establish an improved XGBoost model to predict the breakdown field strength. The GA algorithm automatically selects the optimal parameters during the training process of the XGBoost model. The Pearson correlation coefficient is used to analyze the influencing factors. It can be seen that the thickness and temperature are the key factors affecting the breakdown field strength, while the influence of the masterbatch content, the diameter and relative content of Al(OH)3 micropowder is relatively small. The evaluation index of the commonly used regression model is compared with the proposed model. The coefficient of determination of the model can reach 0.953, and the root-mean-square error and mean absolute error are only 0.361 kV/mm and 0.168, respectively. The results show that the GA-XGBoost model can accurately predict the breakdown fieid strength of the material, which can provide a reference for studying the properties of silicone rubber materials and optimizing the material formulation.
WANGX L, LIUH Y, GUZ Y,et al .Study on the silicone rubber materials and aging problems for transmission lines[J].Chemistry and Adhesion,2024,46(1): 49-52.(in Chinese)
LIUX J, SHENY, JIAF R,et al .Molecular simulation of thermal cracking process of nano-SiO2/silicone rubber composites[J].High Voltage Engineering,2024,50(12): 5397-5405.(in Chinese)
BIM Q, SHUAIY L, DENGR K, et al. Study on the influence of haze parameters on the external insulation characteristics of corona-aged silicone rubber [J]. Journal of Chongqing University of Technology (Natural Science), 2023,37(2):251-259. (in Chinese)
BIM Q, YANGZ P, JIANGT Y, et al. Corona aging characteristics of silicone rubber with different proportions under haze conditions[J]. Journal of Hunan University (Natural Science), 2022, 49(12):167-175. (in Chinese)
[9]
代冲 .低温下硅橡胶绝缘材料电气性能研究[D].北京:华北电力大学,2017.
[10]
DAIC .Study on electrical properties of silicone rubber insulation material at low temperature[D].Beijing:North China Electric Power University,2017.(in Chinese)
HUJ L, TANC, ZHONGR,et al .Influence of the thickness of room temperature vulcanized silicone rubber coatings on the aging characteristics of corona[J].High Voltage Engineering, 2020, 46(4):1328-1336.(in Chinese)
ZHOUY X, HOUF, NIEQ,et al .Temperature effects on electrical tree aging characteristics of silicone rubber[J].High Voltage Engineering,2012,38(10):2640-2646.(in Chinese)
[15]
NANDIA K. Genetic algorithm-based design and development of particle-reinforced silicone rubber for soft tooling process[J]. Materials and Manufacturing Processes, 2013, 28(7):753-760.
[16]
SHAYMAAA Q, SAUDA A, KAREEMA. Grain size effects on the behavior of silicone rubber high voltage power cables using seagull optimization algorithm[J]. The Indonesian Journal of Electrical Engineering and Computer Science, 2023, 32(3):1246-1256.
[17]
CHENZ Q, JIAW, CHENGL, et al. Prediction of breakdown field strength for large-area and multilayer film dielectrics[J]. IEEE Transactions on Dielectrics and Electrical Insulation, 2022, 29: 470-477.
XUX R, GAOY, LIJ,et al .Prediction of thickness-dependent DC breakdown strength of polyethylene nanocomposite estimated with surface potential decay measurement[J].High Voltage Engineering,2022,48(7):2864-2872.(in Chinese)
[20]
LIZ L, DONGY M, WUY,et al .Breakdown performance evaluation and lifetime prediction of XLPE insulation in HVAC cables[J]. Energies, 2024, 17(6): 1337.
WANGH X, LIY, LINH Q,et al. Study on change regularity of DC breakdown characteristics of silicone rubber at low temperature[J]. High Voltage Apparatus, 2019,55(3):103-108.(in Chinese)
General Administration of Quality Supervision,Inspection and Quarantine of the People’s Republic of China,Standardization Administration of the People’s Republic of China. Insulating materials for electrical strength-test methods-Part 1:tests at power frequencies:GB/T 1408.1—2016 [S]. Beijing:Standards Press of China,2016.(in Chinese)
[25]
BIM Q, LIK, ZHANGS Y,et al .Electric field simulation analysis of ICOG corona ring and study on flashover characteristics of bird droppings[J].Electric Power Systems Research,2024,233:110459.
WANGR C, HEY Y, KANGH W,et al .Thermal aging and ultrasonic characteristics of silicone rubber for cable joint insulation[J].High Voltage Engineering,2021,47(9):3181-3188.(in Chinese)
YANGT, HAOJ, CHENX,et al .Study on moisture absorption,power frequency breakdown and gas production characteristics of three-element mixed insulation oil[J]. High Voltage Engineering,2020, 46(11): 4014-4022.(in Chinese)
General Administration of Quality Supervision,Inspection and Quarantine of the People’s Republic of China,Standardization Administration of the People’s Republic of China . Rubber,vulcanized—determination of electrical breakdown strength and voltage resistant at commercial power frequency:GB/T 1695—2005 [S].Beijing:Standards Press of China,2006.(in Chinese)
GAOJ, YANZ M, LIJ Y, et al. The influence mechanism of temperature on the self-recovery characteristics of silicone rubber electrical tree[J]. Journal of Xi’an Jiaotong University, 2022, 56(9): 176-184. (in Chinese)
WANGX, WANGH N, CHENF P,et al .Effect of corona-discharge-accelerated deterioration of silicone grease on the surface properties of silicone rubber insulation[J].High Voltage Engineering,2020,46(11):3977-3985.(in Chinese)
[36]
DASM K, CHAUDHARYA, BRYANA,et al .Rapid screening evaluation of SARS-CoV-2 IgG assays using Z-scores to standardize results[J].Emerging Infectious Diseases, 2020, 26(10): 2501-2503.
[37]
ZHANGL, WANGF L, SUNT,et al .A constrained optimization method based on BP neural network[J].Neural Computing and Applications,2018,29(2): 413-421.
[38]
OMERS G, LIORR K, Approximating XGBoost with an interpretable decision tree [J]. Information Sciences, 2021,572:522-542.
WANGC G, QINH R, YANGC Y, et al. Prediction of frozen soil strength and analysis of influencing factors based on XGBoost algorithm[J/OL]. China Industrial Economics,2024:1-14.(2024-05-06)[2024-06-04]. in Chinese)
WANGL Y, SUNY L, ZHANGG G,et al .Predicting polishing performance of magnetic abrasive and optimizing its preparation process parameters based on Taguchi-GA synergy[J].Journal of Hunan University (Natural Sciences),2024,51(4):43-53.(in Chinese)
[43]
NGUYENV L, SANGW H, KIHAKL, Genetic algorithm hybridized with extreme gradient boosting to predict axial compressive capacity of CCFST columns[J]. Composite Structures, 2021,278:114733.
[44]
TOMALJ H, WELCHW J, ZAMARR H. Robust ranking by ensembling of diverse models and assessment metrics[J]. Journal of Statistical Computation and Simulation, 2023, 93(1):77-102.